At 3:05 AM 12/4/4, Harry Veeder wrote:
>Since it is acceptable to question conservation laws on this forum,
>perhaps CF is possible because the charge on subatomic particles is not
>conserved in all contexts.
Irreverance here, especially amateur irreverence, also sometimes extends to
thermodynamics. Why not? Something might even be learned by just bungling
around.
Second Law Violating Nanochip (SLVN)
GENERAL
The purpose here is to discuss issues regarding the construction of a
Second Law Violating Nanochip (SLVN). The goal is to design a device that
demonstartes that the assumed Second law of Thermodynamics is invalid by
showing that it is possible to extract heat from one of two equal
temperature compartments to increase the temperature of the second
compartment. That is to say, extract kinetic energy from matter in the
first compartment, convert it to electrical energy, and heat the second
compartment via a joule heater.
THE PIEZO-KINETIC APPRAOCH
Let us consider the possiblility of manufacturing a chip with very
thin very small piezoelectric crystals on the surface connected
to integrated fullwave diode bridges. The output of all the tiny bridges
would be collected together, the chip placed into a havy noble gas. Suppose
the chip is placed in a compartment adjacent to which is another
compartment at the same temperature. The chip drives a joule heater in the
second compartment. At
some operating temperature the chip it might be possible to convert kinetic
energy from one compartment to electrical energy, which is then transferred
to the kinetic energy of the second compartment. The difficulty is making
the chip so it will not be destroyed by the operating temperature and the
piezoelectric
crystals small enough in area compared to the size of the impinging gas
molecule, so that the voltage generated by the piezo-compression is
sufficient to make it through the diode bridge, i.e. overcoming the diode
forward bias potential. The peizo must have a small surface area to
prevent the charge being spread over a wide plate, thus reducing the
voltage. Any required energy requirement can be met by utilizing a
sufficient particle energy, or operating temperature. The main
difficulties are achiving a small piezo area, a small integrated fulwave
bridge in the same cross section, and low enough diode forward bias.
For a rough first cut at this assume an operating temp of 300 K.
Since 1 eV = 11,600 K, at 300 K the typical particle in a gas will have an
energy of 300/11,600 eV = .026 eV = .026 * (1.602 x 10^-19 J/eV) = 4.166 x
10^-21 J.
Let's assume we want to charge a capacitor to .3 V. Since E = .5(C)V^2 we
get C = 2E/V^2 = 2*(4.166 x 10^-21 J)/.09 F = 9.76 x 10^-20 F.
Now C = Ke (A/w) (8.85 x 10^-12 F) where Ke is the dielectric constant, A
is the plate area in m, and w is the thickness of the capacitor in m. For
the sake of simplicity and to get scale, let's assume A = w^2, and Ke = 4,
so C = 3.54 x 10^-11 F/m * w. So now w = (9.76 x 10^-20 F)/(3.54 x 10^-11
F/m) = 2.76 x 10^-9 m. The structure size for the device should be in the
range of about 27.6 Angstroms. The atomic radii of Si, O, and Au are 1.46
A, .65 A, and 1.79 A respectively. So 27 A represents a structure about
7-10 atoms across.
However, this assumes a perfectly non-elastic collision every time
(estimate optimistic), yet the kinetic energy of a gas is a distribution,
so many collisions will be more energetic, some much more so (estimate
pessimistic).
So, what does this say? The design is infeasible. The structures are too
small to be practical or functional. The difficulty centers about the need
to focus on a small enough area a sufficient amount of energy to overcome
the forward bias of the diode. The forward bias sets a minimumn voltage
level, which sets a maximum surface area over which the generated charge is
to be distributed. If the forward bias of the diode were zero then there
would be no upper limit to the size of the energy trapping structure, but
like with browian motion, smaller gives more of a result.
What about power? If such a device can be built that works at all, then
there is a very good potential for significant energy production. This is
because, assuming some of the generated energy is returned to stir the gas,
a very large percentage of the molecules will connect with the sides of the
container per second. This means a significant portion of the specific heat
of the gas could be drained off per second. One problem with the chip might
be maintaining balance, not cooling the compartment so much the energy is
not transferred and yet not overheating the chip. But those are much easier
problems.
THE CHARGE-TRANSPORT APPROACH
Having seen some of the difficulties of extracting energy from neutral gas
particles, it is now easier to appreciate the advantages of extracting
energy from an electrolyte. Here, the idea is to use local charge
fluctuations, thus indirectly heat, in an electrolyte to drive the chip.
Similar particle kinetic energies apply, based on temperature, however, the
energy of individual particles (or clusters of particles, or even large
brownian type particles) is expended driving a charged particle to an
electrode, as opposed to driving a neutral particle to a crystal to
generate a piezo electric effect. One adavntage of this approach is that
the operation directly results in electrical energy. Other advantages are
increased efficiency due to less generation from the resulting collisions,
and a reduction in the number of parts to each element on the nano-chip
(only a diode and protective covering is required.)
The method is to build the chip out of vertical diodes separated, i.e.
surrounded by, a lattice of insulating material. Two type of chips could
be built, positve exposed end (PEE) and negative exposed end (NEE) diodes,
as opposed to NP or PN. The diodes would have one end attached to a shared
conducting plane, the other end exposed to the electrolyte. Except for the
conducting plane shared in common, the diodes would be electrically
isolated from each other except through the electrolyte. The face of each
diode would be hardened with a layer of gold.
A PEE chip could be manufactured, for example, by building on top of a
substrate an N doped layer, then a P doped layer, followed by a Gold layer.
This could be followed by cutting a lattice of grooves, leaving a matrix of
small independent diodes, and then filling the groves with an electrolyte
impervious insulating material and then removing the top layer of
insulating material sufficiently to expose the gold contact points but not
the diode material.
The resulting PEE SLVN would look like the following:
EEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEE
HHHHHHHH GGGGGGGG HHHHHHHH GGGGGGGG HHHHHHHH GGGGGGGG HHHHHHHH
HHHHHHHH GGGGGGGG HHHHHHHH GGGGGGGG HHHHHHHH GGGGGGGG HHHHHHHH
HHHHHHHH PPPPPPPP HHHHHHHH PPPPPPPP HHHHHHHH PPPPPPPP HHHHHHHH
HHHHHHHH PPPPPPPP HHHHHHHH PPPPPPPP HHHHHHHH PPPPPPPP HHHHHHHH
HHHHHHHH NNNNNNNN HHHHHHHH NNNNNNNN HHHHHHHH NNNNNNNN HHHHHHHH
HHHHHHHH NNNNNNNN HHHHHHHH NNNNNNNN HHHHHHHH NNNNNNNN HHHHHHHH
HHHHHHHH SSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSS HHHHHHHH
HHHHHHHH SSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSSS HHHHHHHH
HHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHHH
E - electrolyte,
G - gold or other protective conductor,
H - insulating material impervious to electrolyte,
N - N type semiconductor
P - P type semiconductor
S - substrate
A PEE type array and a NEE type array would be placed in an electrolyte
simulataneously. As the random motion of the liquid would bring charges
close,
and then move them away, the induced field in the diodes would cause
current to flow, but primarily in accord with the diode polarity. Moving
the charges away from the exposed diode ends would reduce the kinetic
energy of the electrolyte. A charge would be build up on the exposed end of
the diode which would eventually attract an ion that would be neutralized.
Electrolysis would result. This brings the diode back to a neutral positon
to recycle. Three good things happen: the electrolyte cools, current is
available, and you get electrolysis.
It seems reasonable that this is a workable idea, based upon the problems
and limits of chip miniaturization. One limit is power density, but another
is the fact that molecules bouncing off chips produce electric pulses. The
fact that the electric pulses can be significant at some level of
miniaturization means that impacting molecules are able to generate
voltages in excess of the minimum required to exceed the diode or
transistor bias.
However, to make the design more practical and immediately implementable, a
method is now suggested to overcome the diode bias potential, the PN
barrier potential, and thus increase the maximum size of the
nano-structures required to extract the kinetic energy of the electrolyte.
The method suggested is to capacitively bias the interface, to establish a
field gradient that increases the ion concentration and charge in the
vicinity of the chip surface.
Here is a suggested arrangement:
++++++++++++++ -----------------
IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII
EEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEEE
PPPPPPPPPPPPPP EEEEEEEEEEEEEEEEE NNNNNNNNNNNNNNNNN
IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII
+ - a positive conductive plate
- - a negative conductive plate
I - insulating wall of electrolyte container
E - electrolyte
N - NEE chip
P - PEE chip
The PEE and NEE chips would be electrically connected via a power
extraction device. Such a power extraction device might include a pump of
some kind to move the electrolytic fluid. Such movement, in addition to
assisting in degassification, would actually increase the electrolysis. It
should be noted that such electrolysis resulting strictly from electrolyte
flow, i.e. energy extraction from the the fluid flow, is exactly offset by
pump power requirements. The movement of charge against a field gradient
increases the fluid flow resistance.
The main advantage of this approach is to bias the input potential to the
diode array by jamming a large number of ions of the correct potential
toward the surface of the diode array, thus minimizing the fluctuations in
potential necessary to cause current flow across the diode forward
potential barrier. Note that no energy is extracted from the charged metal
plates which create this bias potential, as they are insulated from the
electrolyte. The + plate drives + ions toward the P type semiconductor,
which positively biases the the P side of the PN junction. This bias allows
the small ionic noise voltage oscillations to generate current across the
junction. If a plus charge is lost to the diode current, this creates a
plus deficit or net negative charge on the surface of the PEE chip. This
charge deficit is made up through electrolyte diffusion, thus heat is
extracted from the diffusion processs itself. If the fluid is flowing then
this process is further enhanced.
Due to the statistical effects of the motion, electrical noise level and
the oscillating potentials in an electrolyte, especially a flowing
electrolyte, can get fairly high compared to the potential a single charged
particle can cause. A kind of macro level thing happens, similar to
brownian motion. It could possibly be enhanced with colloids.
Of course pumping, H2 extraction, etc., are simply practical matters. The
electrical power generated could also be used to heat the theoretically all
important second compartment. The important issue here is hope. If the
Second Law is dead, there is then solid hope for "the" energy solution. A
successful yet very small scale demonstration would be of very great
scientific and practical value.
The present chip technology keeps getting smaller and smaller, and the
diodes and transistors in them work very reliably. Someday nanotechnology
will catch up to thermodynamics and change the rules. The question is,
are close enough today? It appears we are.
References courtesy of Dr. Hal Puthoff who posted them on the vortex-l list:
Yater, Power conversion of energy fluctuations," Phys. Rev. A 10, 1361
(1974); Comments by EerNisse, Phys Rev A 18, 767 (1978); Rebuttal by Yater,
Phys Rev A 20, 623 (1979).
See also articles by Maddox in Nature with titles "Directed motion from
random noise," and "Bringing more order out of noisiness," both in vol 369,
pp. 181 and 271 (1994).
Finally article J. Travis, "Making light work of Brownian motion," Science
267, 1593 (1995).
Regards,
Horace Heffner
